Download A Purine-Pyrimidine Classification Scheme of the Genetic Code

A Purine-Pyrimidine Classification Scheme
of the Genetic Code
Although containing the same information, our new classification scheme of
the genetic code is simpler than the common representation as a threedimensional matrix: it contains just 32 instead of 64 fields. Moreover, it shows
known patterns in the code more clearly than the common scheme. Above all,
with the help of our new scheme we could identify new patterns never seen
before. This gives rise to some speculations about the origin and early
evolution of the genetic code. We hypothesize that coding started in a binary
doublet manner and developed via a quaternary doublet code to our
contemporary quaternary triplet code. Most interestingly, it may be possible
to discover traces of the old binary coding in present-day genomes.
Thomas Wilhelm, Svetlana Nikolajewa
The genetic code specifies how the information contained in the nucleic acids
DNA and RNA is translated into the correct sequence of amino acids building
the highly specific proteins. Up to the
three termination codons UGA, UA(G/A)
(standard code), each nucleotide triplet
stands for exactly one amino acid, the
methionin codon AUG is also the start
codon. The genetic code is comma-free
and non-overlapping. It is usually represented as a three-dimensional matrix in
which the four rows stand for the first
base and the four columns for the second
base. To show the third dimension (the
third base) in the plane figure, each of
the 16 boxes is again divided into four
fields, giving together 64 entries (Fig.1).
For a long time one assumed that the
genetic code is universal for all life forms
on earth. Today there are at least 16
slightly deviating different codes known
(www.ncbi.nlm.nih.gov/Taxonomy/Utils/w
printgc.cgi). However, it is generally believed that all these deviations are later
descendants of the earlier standard code.
Not surprisingly, non-standard codes are
only found in small genomes, nearly all of
them in mitochondria known to have by
far the smallest genomes.
Since the early days of the discovery of
the genetic code non-random patterns
have been searched in the code for providing information about its origin and
early evolution. In 1965 Nirenberg finished his famous project of deciphering
the code. At that time most scientists believed that the code is the result of pure
chance and hence does not need any further evolutionary explanation. Crick [1]
formulated the corresponding “frozen accident” hypothesis which was widely accepted for many years. However, today it
is assumed that at least some hints of possible evolutionary scenarios can be found
in our contemporary code. The top-down
approach, which we are following here,
analyzes patterns in the code and tries to
infer appropriate chemical and selective
forces. The bottom-up approach, on the
other hand, is rooted in biochemistry and
aims at constructing plausible scenarios
for the origin of coding.
It has been appreciated for a long time
that the genetic code assigns similar amino
acids to similar codons. Two different rationales have been presented: first, mutation
and translation error minimization [2], and
second, similar amino acids tend to directly
interact with similar RNA sequences [3]. It
was stated that “the canonical code is at or
very close to a global optimum for error
minimization” [4]. It has also been proposed that instead of the actual codons,
some of their derivatives, such as the anticodons or codon-anticodon duplexes were
the original amino acid binding motifs. It is
also possible that the original amino acid
recognition took place at the tRNA acceptor stem. Szathmary [5] proposed that
amino acid-RNA allocation took place even
BIOforum Europe 06/2004, pp 46–49, GIT VERLAG GmbH & Co. KG, Darmstadt, www.gitverlag.com/go/bioint
Fig. 1: The common representation of the standard genetic code (mRNA triplets in the mRNA reading direction (5'_3')). Shaded regions show codon families.
before the appearance of tRNA. He also
gave a possible evolutionary scenario for
the development of an anticodon hairpin to
a longer structure with an operational code
at the acceptor stem.
However, the first codon position
seems to be correlated with amino acid
biosynthetic pathways and to their evolution as evaluated by synthetic “primordial soup” experiments. The second position is correlated with the hydropathic
properties of the amino acids, and the
degeneracy of the third position could be
related to the molecular weight or size of
the amino acids [6]. Lagerkvist [7] observed that codon families (the amino
acid of a codon family is uniquely determined by the first two nucleotides of a
codon) have a much higher probability to
appear in the left part of the common illustration scheme (cf. Fig. 1). He also
found that “strong” codons (the first two
nucleotides in the codon are G and/or C)
always represent codon families, while
“weak” codons (A and/or U as the first
two nucleotides) never do so. “Mixed”
codons in the right part of the scheme
never represent codon families, whereas
mixed codons in the left part always
stand for a codon family.
The New Classification Scheme
of the Genetic Code
Most amino acid properties show no
clear pattern in the common scheme of
the genetic code. Recently we proposed a
new classification scheme [8 and
www.imb-jena.de/~sweta/genetic_code],
based on a binary purine(1)-pyrimidine(0) coding (Fig. 2). It shows known
regularities more clearly than the common scheme and it even highlights some
new patterns.
There are three possible variants of a
binary coding scheme for the genetic
code: One could group the bases (i) according to base-pairs (A,U = 1, G,C = 0),
(ii) according to keto- and aminobases
(G,U = 1, A,C = 0), and (iii) according to
purines and pyrimidines (A,G = 1, C,U =
0). In such a simplified code eight different binary triplets exist: 000, 001, ..., 111.
Each of these binary triplets represents
eight different codons, e.g. in our coding
scheme 000 stands for CCC, CCU, ...,
UUU. The purine-pyrimidine coding is superior to the other two variants, because
it is the only one that allows the genetic
code to be represented using just four
columns (Fig. 2). The reason for this vast
simplification in our scheme is that for
the third position in each triplet it only
matters if it is a purine or a pyrimidine.
Given the primary purine-pyrimidine
coding, we have again two different possibilities to sort the first two bases per
row: one can use either of the remaining
two binary codings, according to basepairs or according to keto- and
aminobases as a sort criterion inside the
rows. We have chosen the base-pairs for
sorting inside rows, because only this reveals the following regularities of the genetic code: (i) All codon families group
together, i.e. they are not scattered
allover the table. (ii) More importantly,
the codon strength classification directly
corresponds to the columns in our
scheme (cf. Fig. 2). Thus, in the first column the first two bases complementary
pair with 6 hydrogen bonds, in the sec-
Fig. 2: The purine(1)-pyrimidine(0) classification scheme of the standard genetic code. The third base
is given in parenthesis. If there are differences between the standard code and any other code, the
number of deviations from the standard code is indicated. For instance, in the UG(G/A) field, 0/9 indicates that UGG encodes for Trp in all codes, but UGA is not the termination codon in 9 of the 16
non-standard codes. In some bacteria the 21st amino acid, selenocysteine, can also be encoded by
UGA. Shaded regions show codon families. The point in the center indicates the perfect point symmetry corresponding to Halitsky’s family – nonfamily symmetry operation [9]. The thick horizontal
line marks the symmetry axis for codon-anticodon symmetry.
ond and third column with 5, and in the
fourth column with just 4 hydrogen
bonds. For all these reasons our classification scheme of the genetic code is superior to all similiar ones.
Our new scheme shows some fascinating regularities. We can, for instance,
better understand the number of different tRNAs in some organisms. In the simplest case one should expect one tRNA
per coding field in our scheme. Exactly
this happens in the case of vertebrate
mitochondria. It is known that animal
mitochondria contain exactly 22 different tRNAs. In vertebrate mitochondria
UA1 and AG1 are stop codons. Thus
there are exactly 22 fields for amino
acids left: the 8 codon families plus 14
remaining fields. Interestingly, the 22 tRNAs in animal mitochondria correspond
1:1 to these 22 fields.
The amino acids of the nine "strong
groups" (mutually evolutionary conserved,
based on the alignment score matrix
PAM250, cf. http://bioinfolab.unl.edu/emlab/documents/clustalx_doc/clustalw.txt)
very closely group together in our scheme,
more closely than in the standard scheme.
That means neighboring amino acids in
our schme have a higher probability to be
aligned to each other in genome comparisons than neighboring amino acids in the
standard scheme.
Our new scheme also led us to detect
hitherto unknown regularities of amino
acid properties in the genetic code.
Jungck [10] collected 15 different measures of amino acid properties. For all of
these we arranged a table with 8 rows
and 4 columns corresponding to our
scheme. Amazingly, the column sums of
nearly all measures are perfectly correlated to the corresponding codon-anticodon binding strength. For instance, the
first column harbours more polar amino
acids, the last column less polar ones and
the mixed codon fields are in between.
Similarly, the bulkiness and the specific
volume increases continuously from the
first to the last column.
Evolution of the Genetic Code
The observed regularities inspire to
some speculations about the early evolution of the genetic code. Thus the strong
correlation between amino acid properties and codon strength implies that the
first two positions together (and not the
second position alone as speculated by
others) must have been important for the
amino acid – codon assignment in the
early evolution of the code. It therefore
also could be that just the first two nucleotides of a codon (or anticodon) show
specific binding affinity to the corre-
sponding amino acid (maybe important
in the process of the code formation).
Nowadays one assumes that "the code
probably underwent a process of expansion from relatively few amino acids to the
modern complement of 20" [11]. Can we
find some hints in our scheme indicating
coding of less than 20 amino acids in ancient times? Indeed, there is a high redundancy for each second row. This gives rise
to the speculation that in the early days of
code evolution just the first two bases of
the triplet were coding. The reading
frame, however, arguably always comprised three letters. In any way, a quaternary doublet can encode at most 16 amino
acids, or 15 plus one termination codon
(some bacteria exist that do not possess
any stop codon). In this context it is interesting to note that Asn, Gln, Met, Trp, and
Tyr seem to be newer amino acids.
Since the discovery of the genetic code it
is speculated that the first genetic material
contained only a single base-pairing unit
[1]. Recently, for the first time a ribozyme
was found composed of only one purine
and one pyrimidine [12]. Assuming a binary doublet code, it is tempting to speculate which four amino acids, one per two
consecutive rows, were the first encoded
ones. In the first two rows Ser seems to be
the oldest amino acid, and in the third and
fourth row Ala. The 01-rows obviously contain no really old amino acid while the 11rows contain more than one: Gly, Asp, Glu.
However, Gly is biochemically built from
Ser, so Ser can be assumed as prior. It
could be that in the beginning of nucleic
acid – amino acid assignment Asp and Glu
competed for the 11-doublet. Of course,
code transfer from one amino acid to another might also have occurred.
Conclusions
We have found a concise scheme for the
genetic code that is superior to similar
schemes for different reasons. It shows
nice patterns and symmetries and even
so far unknown regularities in the code.
We are now studying the fascinating
question whether we still can find traces
of doublet coding or even binary coding
in contemporary genomes.
References are available from the authors.
Dr. Thomas Wilhelm
Svetlana Nikolajewa
Theoretical Systems Biology
Institute of Molecular Biotechnology
Beutenbergstr. 11
07745 Jena, Germany
[email protected]
[email protected]